
Engineering Cohort · 2026A client intranet launch is two days out. Lighthouse just tanked, axe flagged the new hero, and the strategist wants a content model change—what ships?
Each module is a real call from Local Wisdom client engineering, stripped down to the decision you'd actually have to make before the PR ships.
Module 01
Read the brief, separate must-fix from nice-to-have before you open the editor.
Module 02
Translate a late Figma revision into a token or component decision—not a one-off.
Module 03
Pick the right fix for a contrast regression and a focus-trap inside the new modal.
Module 04
Trade off image strategy, JS code-split, and font loading to recover Lighthouse.
Module 05
Negotiate a late CMS schema change with strategy without blocking the demo.
Bonus
A growing library of patterns other Local Wisdom devs have shipped across client builds.
Before you touch a branch, understand the build. Look for the regressions and asks that will shape what actually ships on Thursday.
Situation: The homepage hero landed on main overnight with a fullscreen animated pattern. QA, accessibility, and strategy all surfaced concerns within the same hour. You decide what gets fixed, what gets reverted, and what waits for the post-launch sprint.
Lighthouse performance dropped from 94 → 62 after last night's hero merge
axe-core flagged 4 serious violations on the new homepage pattern
Strategist just requested a 'Region' field on the Story content model—two days before demo
💡Engineer Tip: A red build isn't a list of bugs—it's a prioritization problem. Triage by user impact and demo risk, not by what shouts loudest in Slack.
You have one working day before the staging cut. Pick the two actions you'd take first.
0/2 selected
A teammate opened the hero PR for review. Click the highlighted regions to find what's solid—and what blocks merge.
Three calls senior LW engineers make every week.
Notes from teammates who ran the training in the last two cohorts.
"Ran the a11y module the morning of a PR review. Caught myself asking the focus-trap question before approving the merge—exactly the way the training framed it."
Priya R.
Senior Engineer · 4 yrs at LW
"The content-model pushback scenario is the one. Every new engineer should run it before they ever touch a Sanity schema two days from a demo."
Marcus T.
Engineering Lead
"I used to silently fix juniors' PRs. Now I leave the review note from the training instead. They level up twice as fast and our patterns actually spread."
Elena K.
Principal Front-End
18
Engineers in cohort
Quarterly
Refresh cadence
Ringoes, NJ
Home base
You've completed Local Wisdom's Engineering Immersive. You have the instincts to triage a red build, defend the user experience, and ship the call that makes the rest of the team faster.
Fix what blocks real people first. Lighthouse follows once the experience is sound.
A red build is a prioritization problem. Sort by user impact and demo risk, not Slack volume.
Reach for Radix/shadcn before hand-rolling focus, dialogs, or popovers. Get a11y right by default.
Clients keep us in the room because our builds hold up under load and audit.
A site everyone can use travels further than one that looks great for a few.
Strong patterns and clear PR reviews keep the whole pod shipping in the same direction.
Bring one of these scenarios to your next engineering standup. Run it live with the pod. The reps are the point.
•Engineering Training · Ringoes, NJ · 2026